A structured evaluation of regression models for predicting CO2 concentration from plasma emission spectra
نویسندگان
چکیده
This study is an extensive comparison of the predictive performance a bagging neural network (BANN), partial least squares (PLS) regressor, and kernel-based, nonlinear PLS (KPLS) given experimentally obtained CO2-N2-Ar plasma emission spectra. The spectra, 3,62,31 in total, were from controlled gas mixtures with varying CO2 concentrations fed to stripline split-ring resonator microplasma source recorded UV-NIR, 2 nm resolution spectrometer. regression methods’ dependence on (i) number observations used training, (ii) preprocessing steps, (iii) feature selection (in this case wavelength ranges) was evaluated by training testing 60–66 models per method, each unique combination aforementioned configuration options, trained 4 times random train-test splits. To compare custom metric that compounds R2, Pearson correlation, weighted root mean squared relative error used. results show BANN outperform both KPLS, reaching peak score 0.873, others getting 0.561 0.581, respectively, using (− ∞ , 1] metric. top performing model without any or preprocessing, these steps were, however, required for best KPLS models. In wider perspective, BANNs are not only suitable as in-place replacements PLS-based methods, but increase prediction accuracy low spectra such extent they offer modelling previously unattainable, information contained datasets.
منابع مشابه
a study on thermodynamic models for simulation of 1,3 butadiene purification columns
attempts have been made to study the thermodynamic behavior of 1,3 butadiene purification columns with the aim of retrofitting those columns to more energy efficient separation schemes. 1,3 butadiene is purified in two columns in series through being separated from methyl acetylene and 1,2 butadiene in the first and second column respectively. comparisons have been made among different therm...
Investigating the Differences in CO2 Emission in the Transport Sector Across Iranian Provinces: Evidence from a Quantile Regression Model
Improving the quality of the environment is a desired goal of any economy. In this regard, given the important role of the transportation sector in the emission of pollutants, policy makers should focus on identifying the factors affecting CO2 emissions in the transport sector. This study examines the factors affecting the differences of CO2 emissions in the transportation sector in Iranian pro...
متن کاملA new method for Eosin preparation and examination of its absorption and emission spectra
Eosin is one of the original dyes of xanthenes derivatives. It is also one of the main derivatives of fluorescein and has brilliant fluorescence. Eosin disodium salt called "Eosin Y" is the most useful form of this compound. In common procedures, eosin is usually obtained by the action of bromine on fluorescein, which has not good yield and purity degree. In this research, it is prepared by the...
متن کاملa new approach to credibility premium for zero-inflated poisson models for panel data
هدف اصلی از این تحقیق به دست آوردن و مقایسه حق بیمه باورمندی در مدل های شمارشی گزارش نشده برای داده های طولی می باشد. در این تحقیق حق بیمه های پبش گویی بر اساس توابع ضرر مربع خطا و نمایی محاسبه شده و با هم مقایسه می شود. تمایل به گرفتن پاداش و جایزه یکی از دلایل مهم برای گزارش ندادن تصادفات می باشد و افراد برای استفاده از تخفیف اغلب از گزارش تصادفات با هزینه پائین خودداری می کنند، در این تحقیق ...
15 صفحه اولذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Spectrochimica Acta Part B: Atomic Spectroscopy
سال: 2022
ISSN: ['0584-8547', '1873-3565']
DOI: https://doi.org/10.1016/j.sab.2022.106467